Cross‐attractor transforms: Improving forecasts by learning optimal maps between dynamical systems and imperfect models
Biased, incomplete numerical models are often used for forecasting states of complex dynamical systems by mapping an estimate of a “true” initial state into model phase space, making a forecast, and then mapping back to the “true” space. While advances have been made to reduce errors associated with model initialization and model forecasts, we lack a general framework for discovering optimal mappings between “true” dynamical systems and model phase spaces. Here, we propose using a data‐driven approach to infer these maps. Our approach consistently reduces errors in the Lorenz‐96 system with an imperfect model constructed to produce significant model errors compared to a reference configuration. Optimal pre‐ and post‐processing transforms leverage “shocks” and “drifts” in the imperfect model to make more skillful forecasts of the reference system. The implemented machine learning architecture using neural networks constructed with a custom analog‐adjoint layer makes the approach generalizable across applications.
document
https://n2t.net/ark:/85065/d7rr23n2
eng
geoscientificInformation
Text
publication
2016-01-01T00:00:00Z
publication
2025-02-28T00:00:00Z
<span style="font-family:Arial;font-size:10pt;font-style:normal;" data-sheets-root="1">Copyright author(s). This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.</span>
None
OpenSky Support
UCAR/NCAR - Library
PO Box 3000
Boulder
80307-3000
name: homepage
pointOfContact
OpenSky Support
UCAR/NCAR - Library
PO Box 3000
Boulder
80307-3000
name: homepage
pointOfContact
2025-07-10T19:54:10.648025